README.md

metadata

language:
  - ar
task_categories:
  - text-to-speech
pretty_name: Mimir Auto Tokenized SNAC

Mimir Auto Tokenized SNAC

Tokenized TTS training dataset derived from the paired audio/text dataset.

Rows have exactly three columns:

• inputs: 8 streams [text_stream, audio_codebook_0, ..., audio_codebook_6]
• labels: same shape as inputs, with -100 for text/prompt positions and real targets only for audio tokens
• attentionmask: sequence mask, 1 for valid positions